Regioselective coupling of pentafluorophenyl substituted alkynes: Mechanistic insight into the zirconocene coupling of alkynes and a facile route to conjugated polymers bearing electron-withdrawing pentafluorophenyl substituents

Article abstract of DOI:10.1021/ja0209161

The reaction of Cp₂ZrCl₂ with 2 equiv of BuLi at -78 °C, followed by the addition of an unsymmetrical tetra- or pentafluorophenyl substituted alkyne R¹C≡CAr_f (R¹, Ar_f = (CH₂)

Full text of DOI:10.1021/ja0209161

Published on Web 03/15/2003
Regioselective Coupling of Pentafluorophenyl Substituted
Alkynes: Mechanistic Insight into the Zirconocene Coupling
of Alkynes and a Facile Route to Conjugated Polymers
Bearing Electron-Withdrawing Pentafluorophenyl Substituents
Samuel A. Johnson, Feng-Quan Liu, Min Chul Suh, Stefan Zu¨rcher, Markus Haufe,
Shane S. H. Mao, and T. Don Tilley*
Contribution from the Department of Chemistry, UniVersity of California, Berkeley,
Berkeley, California 94720-1460
Received July 3, 2002
Abstract: The reaction of Cp2ZrCl2 with 2 equiv of BuLi at -78 °C, followed by the addition of an
unsymmetrical tetra- or pentafluorophenyl substituted alkyne R1CtCArf (R1, Arf ) (CH2)4Me, p-C6F4H;
Me, p-C6F4H; Ph, C6F5), resulted in regioselective couplings of these alkynes to zirconacyclopentadienes
in which the Arf substituents preferentially adopt the 3,4-positions (ââ) of the zirconacyclopentadiene ring.
With Cp2Zr(py)(Me3SiCtCSiMe3) as the zirconocene reagent, the couplings could be carried out at room
temperature; however, at higher temperatures significant quantities of the 2,4-fluoroaryl substituted (Râ)
isomers were also formed. None of the conditions employed produced the 2,5-fluoroaryl substituted (RR)
isomers. These fluoroaryl-substituted zirconacyclopentadienes were readily converted to butadienes via
reactions with acids. The zirconacyclopentadiene Cp2ZrC4-2,5-Ph2-3,4-(C6F5)2, which resulted from the
coupling of PhCtC(C6F5), was converted to the corresponding thiophene by reaction with S2Cl2, and to an
arene by reaction with MeO2CCtCCO2Me/CuCl. Mechanistic studies on zirconocene couplings of
(p-CF3C6H4)CtC(p-MeC6H4) indicate that the observed regioselectivities are determined by an electronic
factor that controls the orientation of at least one of the two alkynes as they are coupled. Additionally,
these studies suggest an unsymmetrical transition state for the zirconocene coupling of alkynes, and this
is supported by DFT calculations. The reaction of [(C6F5)CtCCH2]2CH2 with Cp2Zr(py)(Me3SiCtCSiMe3)
resulted in a zirconacyclopentadiene in which the pentafluorophenyl substituents have been forced into
the 2,5-positions (RR). Zirconocene coupling of the diyne (C6F5)CtC-1,4-C6H4-CtC(C6F5) provided a
route to conjugated polymers bearing electron-withdrawing pentafluorophenyl groups.
Introduction
A strategy for obtaining materials with improved electron-
transport properties is the replacement of hydrogen by fluorine
in the monomer units.8-14 The electronegative fluorine substit-
uents lower the energy of the LUMOs in these polymers and
provide increased kinetic stability by virtue of the strength of
the C-F bonds and the steric protection offered by the larger
radius of fluorine vs hydrogen.15 Often, the syntheses of these
polymers are neither facile nor high yielding. Furthermore,
despite the current interest in fluorinated polymers, few
Conjugated oligomers and polymers have attracted interest
for their ability to replace inorganic components in electronic
devices such as light-emitting diodes (LEDs)1,2 and thin-film
transistors.3 The majority of materials used in these electronic
devices are good positive charge (p-type) semiconductors, such
as polythiophene,4 poly-p-phenylene, and poly-p-phenyl-
vinylene.5 In general, these polymers possess relatively high
energy LUMOs, which results in poor negative charge (n-type)
conduction and limits their utility in electronic devices.6 The
development of n-type organic semiconductors could result in
improved efficiencies for LEDs7 and new applications for
organic semiconductors.
(8) Sakamoto, Y.; Komatsu, S.; Suzuki, T. J. Am. Chem. Soc. 2001, 123, 4643-
4644.
(9) Heidenhain, S. B.; Sakamoto, Y.; Suzuki, T.; Miura, A.; Fujikawa, H.;
Mori, T.; Tokito, S.; Taga, Y. J. Am. Chem. Soc. 2000, 122, 10 240-
10 241.
(10) Sakamoto, Y.; Suzuki, T.; Miura, A.; Fujikawa, H.; Tokito, S.; Taga, Y. J.
Am. Chem. Soc. 2000, 122, 1832-1833.
(11) Babudri, F.; Cardone, A.; Chiavarone, L.; Ciccarella, G.; Farinola, G. M.;
Naso, F.; Scamarcio, G. Chem. Commun. 2001, 1940-1941.
(12) Brooke, G. M.; Drury, C. J.; Bloor, D.; Swann, M. J. J. Mater. Chem.
1995, 5, 1317-1320.
(1) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay,
K.; Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539-
541.
(2) Mitschke, U.; Bauerle, P. J. Mater. Chem. 2000, 10, 1471-1507.
(3) Kraft, A. ChemPhysChem 2001, 2, 163-165.
(4) McCullough, R. D. AdV. Mater. 1998, 10, 93-116.
(5) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of
Conducting Polymers; 2nd ed.; Marcel Dekker: New York, 1997.
(6) Laquindanum, J. G.; Katz, H. E.; Dodabalapur, A.; Lovinger, A. J. J. Am.
Chem. Soc. 1996, 118, 11 331-11 332.
(7) Lee, Y.-Z.; Chen, X.; Chen, S.-A.; Wei, P.-K.; Fann, W.-S. J. Am. Chem.
Soc. 2001, 123, 2296-2307.
(13) Kiebooms, R.; Adriaensens, P.; Vanderzande, D.; Gelan, J.; Swann, M. J.;
Bloor, D.; Drury, C. J.; Brooke, G. M. Synth. Met. 1997, 84, 189-190.
(14) Kiebooms, R.; Adriaensens, P.; Vanderzande, D.; Gelan, J.; Swann, M. J.;
Bloor, D.; Drury, C. J.; Brooke, G. M. Macromolecules 1996, 29, 5981-
5989.
(15) Facchetti, A.; Deng, Y.; Wang, A.; Koide, Y.; Sirringhaus, H.; Marks, T.
J.; Friend, R. H. Angew. Chem., Int. Ed. 2000, 39, 4547-4551.
9
10.1021/ja0209161 CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 4199-4211
4199

A R T I C L E S
Johnson et al.
Scheme 1
conjugated materials have been prepared that contain the
considerably more electron-withdrawing perfluorophenyl group
as a substituent,16,17 and simple methodologies for the prepara-
tion of perfluoroaryl-substituted conjugated polymers and oli-
gomers are not well developed.
Our studies have focused on the use of zirconocene synthons
for the coupling of alkynes to produce zirconacyclopentadiene-
containing macrocycles,18-24 oligomers25,26 and polymers.27-29
The zirconacyclopentadiene moiety can then be converted into
other derivatives such as dienes,30 arenes,31-36 cyclopenta-
dienes37,38 and heterocycles39 such as thiophenes,40,41 siloles,42,43
phospholes,44 germoles,27 pyrroles,45 and thiophene oxides.26 The
formation of macrocycles by the zirconocene coupling of
alkynes requires a substituted diyne that couples in a regio-
selective manner, and reversible diyne coupling reactions. To
date, the only alkyne substituent that has fulfilled these roles
are silyl groups, which couple exclusively into the 2,5 (RR)
positions of the zirconacyclopentadiene ring.46-48 Both the
selectivity and the reversibility of the coupling of silyl substi-
tuted alkynes may be due to the size of these groups; however,
little is known about the role of electronic effects in the coupling
of alkynes by the zirconocene fragment.49 The availability
of additional substituents which could direct regioselective
alkyne coupling would increase the scope of this reaction for
the synthesis of functionalized conjugated polymers and mac-
rocycles.
This paper describes the effects of electron-withdrawing
perfluoroaryl groups on the zirconocene coupling of alkynes,
and demonstrates that these substituents direct the regio-
chemistry of the coupling such that the perfluoroaryl groups
selectively adopt the 3,4-positions (ââ) of the zirconacyclopen-
tadiene ring. A study of the mechanism of this reaction indicates
that the coupling of alkynes by the zirconocene fragment occurs
via an unsymmetrical transition state. Preliminary investigations
indicate that this â-directing effect may be used to generate
pentafluorophenyl-substituted conjugated polymers, which may
have applications as electron-transport materials.
(16) Yoshimura, T.; Masuda, T.; Higashimura, T.; Okuhara, K.; Ueda, T.
Macromolecules 1991, 24, 6053-6058.
(17) Yoshimura, T.; Masuda, T.; Higashimura, T.; Ishihara, T. J. Polym. Sci.,
Part A: Polym. Chem. 1986, 24, 3569-3572.
(18) Mao, S. S. H.; Liu, F.-Q.; Tilley, T. D. J. Am. Chem. Soc. 1998, 120, 1193-
1206.
(19) Schafer, L. L.; Nitschke, J. R.; Mao, S. S. H.; Liu, F.-Q.; Harder, G.; Haufe,
M.; Tilley, T. D. Chem. Eur. J. 2002, 8, 74-83.
(20) Nitschke, J. R.; Tilley, T. D. J. Am. Chem. Soc. 2001, 123, 10 183-10 190.
(21) Nitschke, J. R.; Tilley, T. D. Angew. Chem., Int. Ed. 2001, 40, 2142-
2145.
(22) Schafer, L. L.; Tilley, T. D. J. Am. Chem. Soc. 2001, 123, 2683-2684.
(23) Nitschke, J. R.; Zuercher, S.; Tilley, T. D. J. Am. Chem. Soc. 2000, 122,
10 345-10 352.
Results and Discussion
Regiochemistry of the Zirconocene Coupling of Fluoro-
phenyl-Substituted Alkynes. The fluorophenyl-substituted
alkynes used in this study (1a-d) were synthesized in moderate
to high yields by the Pd(PPh3)4/CuI-catalyzed coupling of the
appropriate fluoroaryl iodide with a terminal alkyne. Reaction
of alkynes 1a-c with Negishi’s zirconocene synthon50 at low
temperature followed by warming to room temperature over 6
h afforded zirconacyclopentadienes 2a-c in high yields, as
shown in Scheme 1. For 1a and 1b, one of the three possible
zirconacyclopentadiene isomers was formed almost exclusively
(>97%); however, the phenyl-substituted alkyne 1c produced
the symmetrical isomer 2c(ââ) in 85% yield and the unsym-
metrical coupling product 2c(Râ) in 15% yield. The selectivities
of these reactions were determined by 1H, 13C{1H} and 19F{1H}
NMR spectroscopies of the crude reaction products.
(24) Liu, F.-Q.; Harder, G.; Tilley, T. D. J. Am. Chem. Soc. 1998, 120, 3271-
3272.
(25) Suh, M. C.; Jiang, B.; Tilley, T. D. Angew. Chem., Int. Ed. 2000, 39, 2870-
2873.
(26) Jiang, B.; Tilley, T. D. J. Am. Chem. Soc. 1999, 121, 9744-9745.
(27) Lucht, B. L.; Buretea, M. A.; Tilley, T. D. Organometallics 2000, 19, 3469-
3475.
(28) Lucht, B. L.; Mao, S. S. H.; Tilley, T. D. J. Am. Chem. Soc. 1998, 120,
4354-4365.
(29) Mao, S. S. H.; Tilley, T. D. Macromolecules 1997, 30, 5566-5569.
(30) Lucht, B. L.; Tilley, T. D. Chem. Commun. 1998, 1645-1646.
(31) Takahashi, T.; Xi, Z.; Yamazaki, A.; Liu, Y.; Nakajima, K.; Kotora, M. J.
Am. Chem. Soc. 1998, 120, 1672-1680.
(32) Takahashi, T.; Kotora, M.; Xi, Z. J. Chem. Soc., Chem. Commun. 1995,
361-362.
(33) Takahashi, T.; Hara, R.; Nishihara, Y.; Kotora, M. J. Am. Chem. Soc. 1996,
118, 5154-5155.
(34) Takahashi, T.; Li, Y.; Stepnicka, P.; Kitamura, M.; Liu, Y.; Nakajima, K.;
Kotora, M. J. Am. Chem. Soc. 2002, 124, 576-582.
(35) Takahashi, T.; Ishikawa, M.; Huo, S. J. Am. Chem. Soc. 2002, 124, 388-
389.
(36) Takahashi, T.; Tsai, F.-Y.; Li, Y.; Nakajima, K.; Kotora, M. J. Am. Chem.
Soc. 1999, 121, 11 093-11 100.
(37) Kotora, M.; Xi, C.; Takahashi, T. Tetrahedron Lett. 1998, 39, 4321-4324.
(38) Takahashi, T.; Xi, C.; Kotora, M. Chem. Commun. 1997, 2069-2070.
(39) Takahashi, T.; Li, Y.; Ito, T.; Xu, F.; Nakajima, K.; Liu, Y. J. Am. Chem.
Soc. 2002, 124, 1144-1145.
To determine the connectivity in 2a-b the corresponding
hydrolyzed compounds 3a and 3b were synthesized as shown
1
in Scheme 2. The H NMR spectra of 3a,b contain triplet and
(40) Fagan, P. J.; Nugent, W. A. J. Am. Chem. Soc. 1988, 110, 2310-2312.
(41) Fagan, P. J.; Nugent, W. A.; Calabrese, J. C. J. Am. Chem. Soc. 1994,
116, 1880-1889.
quartet peaks at δ 5.56 (3JHH ) 7.2 Hz) and 5.42 (3JHH ) 6.8
Hz) for the vinyl protons coupled to the methylene and methyl
groups, respectively, which indicates that the tetrafluorophenyl
groups are in the 3,4-positions of the precursor zirconacyclo-
pentadienes. In the case of 2c(ââ), which was isolated by
crystallization from diethyl ether, X-ray crystallography con-
clusively established the molecular structure (Figure 1). This
compound crystallizes with one equiv of ether. Due to disorder
(42) Tamao, K.; Uchida, M.; Izumizawa, T.; Furukawa, K.; Yamaguchi, S. J.
Am. Chem. Soc. 1996, 118, 11 974-11 975.
(43) Uchida, M.; Izumizawa, T.; Nakano, T.; Yamaguchi, S.; Tamao, K.;
Furukawa, K. Chem. Mater. 2001, 13, 2680-2683.
(44) Hay, C.; Hissler, M.; Fischmeister, C.; Rault-Berthelot, J.; Toupet, L.;
Nyulaszi, L.; Reau, R. Chem. Eur. J. 2001, 7, 4222-4236.
(45) Nakamoto, M.; Tilley, T. D. Organometallics 2001, 20, 5515-5517.
(46) Erker, G.; Zwettler, R. J. Organomet. Chem. 1991, 409, 179-188.
(47) Erker, G.; Zwettler, R.; Krueger, C.; Hyla-Kryspin, I.; Gleiter, R. Organo-
metallics 1990, 9, 524-530.
(48) Erker, G.; Zwettler, R.; Krueger, C.; Schlund, R.; Hyla-Kryspin, I.; Gleiter,
R. J. Organomet. Chem. 1988, 346, C15-C18.
(49) Negishi, E.-I.; Takahashi, T. Acc. Chem. Res. 1994, 27, 124-130.
(50) Negishi, E.; Cederbaum, F. E.; Takahashi, T. Tetrahedron Lett. 1986, 27,
2829-2832.
9
4200 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
Scheme 2
Figure 1. ORTEP depiction of the solid-state molecular structure of 2c(ââ).
Hydrogen atoms and the cocrystallized ether molecule are omitted for clarity.
Ellipsoids are drawn with 50% probabilites. Only Zr(1) and F(1)-F(10)
were refined anisotropically. Selected bond lengths(Å), bond angles(°):
Zr1-C11, 2.281(7); Zr1-C14, 2.278(6); C11-C12, 1.355(8); C12-C13,
1.488(8); C13-C14, 1.333(9); Zr1-C11-C15, 129.5(5); Zr1-C14-C33,
130.0(5).
in the cocrystallized solvent and Cp rings, only the zirconium
and the fluorine atoms were modeled anisotropically.
The selectivity of the alkyne coupling reaction was found to
depend somewhat on the reaction conditions. When the reaction
mixtures of either 1a or 1b with Negishi’s zirconocene synthon
in THF were warmed quickly from -78 °C, up to 12% of the
Râ products were formed, as determined by integration of the
vinylic protons in the 1H NMR spectra of the hydrolyzed
samples. When Cp2Zr(py)(Me3SiCtCSiMe3)23,51 was used as
the zirconocene synthon in toluene at 90 °C, up to 30% of the
Râ isomer was formed, but no detectable amount of the RR
isomer was ever observed. As noted earlier, the coupling of the
phenyl-substituted alkyne 1c was slightly less selective than
those of the alkyl-substituted alkynes (Scheme 1). It is also
notable that the reaction of alkyne 1c with Cp2Zr(py)(Me3SiCt
CSiMe3) is slow at room temperature, whereas alkynes 1a and
1b react within seconds.
The selectivity found in the coupling of the fluoroaryl/alkyl-
substituted alkynes is very different from that found for
nonfluorinated aryl/alkyl-substituted alkynes. For example,
PhCtC(CH2)4CH3 reacts with Negishi’s zirconocene synthon
to produce a coupling product with both phenyl groups in the
R position (15%) and the Râ coupled product (85%), but none
of the product with ââ phenyl substitution.29 The selectivities
observed for the couplings of pentafluorophenyl-substituted
alkynes are also different from those observed in CpCo(CO)2-
mediated alkyne couplings.52,53
Zirconocene Coupling Routes to Fluorinated Derivatives.
Besides hydrolysis, two other transformations of zirconacycle
2c(ââ) were carried out. The reaction of 2c(ââ) with S2Cl2 and
dimethyl acetylenedicarboxylate gave thiophene 4c and arene
5c, respectively, in high yields (Scheme 2).32,41 Single crystals
of 4c suitable for X-ray analysis were grown from an aceto-
nitrile/cyclohexane solution. Its structure is shown in Figure 2.
The UV-vis spectrum of 4c exhibited a λmax value of 320 nm,
which is slightly blue-shifted relative to that for 2,5-diphenyl-
3,4-propanediyl-thiophene (330 nm).25 This may be due to a
smaller degree of conjugation for 4c, which may result from a
larger deviation from planarity due to the larger size of the
Figure 2. ORTEP diagram of one of the crystallographically independent
molecules of thiophene 4c. Hydrogen atoms and the cocrystallized cyclo-
hexane molecule are omitted for clarity. Thermal ellipsoids are drawn with
50% probabilities.
pentafluorophenyl substituents compared to the 3,4-propanediyl
group. This hypothesis is supported by the solid-state molecular
structure of 4c which exhibits rings that are not coplanar.
Mechanistic Study. To determine the relative importance of
electronic and steric effects in the regioselective coupling of
fluoroarylalkynes, the zirconocene coupling of p-tolyl-p-(tri-
fluoromethylphenyl)ethyne (1d) was examined. With the electron-
withdrawing group in the distal para position, the influence of
steric effects during the formation of the zirconacycle should
be minimized. The coupling reactions were carried out by using
either Negishi’s zirconocene synthon at low temperature or
Cp2Zr(py)(Me3SiCtCSiMe3) at room temperature. In the
absence of either electronic or steric effects, a 1:2:1 ratio (RR:
Râ:ââ) for the three possible isomers would be expected. In
1
fact, the H and 19F{1H} NMR spectra indicate that the ââ
isomer and the unsymmetrical Râ isomer were formed as the
major products in a 1:1 ratio. This product distribution was not
strongly affected by reaction conditions; however, the room-
temperature coupling using Cp2Zr(py)(Me3SiCtCSiMe3) pro-
duced a small amount of the RR isomer, which accounted for
5% of the coupling product (5% RR, 45% Râ, 50% ââ). The
two major products could not be separated to conclusively
identify the symmetrical complex as the ââ isomer, so further
reactions were carried out to ascertain this unequivocally.
Hydrolysis of a 1:1 mixture of the symmetrical and unsym-
metrical zirconacycles (2d(ââ) and 2d(Râ)) gave a mixture of
two dienes (3d(ââ) and 3d(Râ)), which was partially separated
by column chromatography. The 2D C-H correlation NMR
spectra (HMQC, optimized for short- and long-range coupling)
and 2D 1H-NOESY NMR-experiments of the symmetrical
(51) Rosenthal, U.; Ohff, A.; Baumann, W.; Tillack, A. Z. Anorg. Allg. Chem.
1995, 621, 77-83.
(52) Dickson, R. S.; Tailby, G. R. Aust. J. Chem. 1970, 23, 1531-1544.
(53) Dickson, R. S.; Fraser, P. J. Aust. J. Chem. 1970, 23, 2403-2411.
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4201

A R T I C L E S
Johnson et al.
Scheme 3
compound confirmed that it is the ââ isomer (see the Supporting
Information). The preferential formation of the ââ and Râ
isomers provides clear evidence that there is a strong electronic
effect in the selectivity of the zirconocene alkyne coupling
reaction.
example, in this case the incoming alkyne orients itself such
that the more electron-withdrawing p-trifluoromethylphenyl
group occupies the â-position in the zirconacyclopentadiene
product (mechanism B in Scheme 3). These mechanisms only
differ in the transition states by which the Râ isomer forms. It
should be noted that these are limiting cases, and it is possible
that with some alkynes both effects could operate to some extent.
By determining the regiochemistries of reactions of fluoroaryl-
substituted alkynes with a variety of zirconacyclopropene
complexes, it should be possible to distinguish between mech-
anisms A and B. However, zirconacyclopropene complexes of
the type Cp2Zr(RCtCR′) have rarely been observed and in
general are not isolable.59 An alternative approach involves the
transient generation of such a species, which may then be
trapped by an added fluoroaryl-substituted alkyne. The micro-
scopic reverse of alkyne coupling by zirconocene (the elimina-
tion of an alkyne from a zirconacyclopentadiene) is presumed
to generate a zirconacyclopropene intermediate,60 but this
reaction has been observed only in the case of zirconacyclo-
pentadienes bearing bulky substituents in the R positions, such
as trimethylsilyl or tert-butyl.49,56,60 However, to further probe
the regioselectivity observed in the coupling of 1d we sought
to observe evidence for reversibility in other zirconacyclo-
pentadienes using relatively high reaction temperatures.
These results also indicate that a stepwise mechanism is
operative. It is believed54 that the initial reaction steps of
zirconocene alkyne coupling involve the formation of a transient
16-electron alkyne complex,55 such as the one shown on the
left side of Scheme 3. This transient zirconacyclopropene then
rapidly inserts a second equiv of alkyne 1d to form the observed
zirconacyclopentadiene products (2d(ââ) and 2d(Râ)). The
presence of only two isomers in the product mixture (2d(ââ)
and 2d(Râ)) implies that the transition state of the C-C bond-
forming step does not contain two equally activated, metal-
bound alkynes, as is often assumed in the coupling of alkynes
by zirconocene.56-58 One possible explanation for the observed
product distribution is that in the unobserved intermediate
monoalkyne complex Cp2Zr[(MeC6H4)CC(C6H4CF3)], the Zr-C
bond associated with the more electron-withdrawing p-tri-
fluoromethylphenyl substituent is significantly more reactive
toward the insertion of a second equiv of alkyne. This reaction
pathway is labeled mechanism A in Scheme 3. A second
possible explanation for the selectivity in the coupling of alkyne
1d involves comparable reactivity for the two Zr-C bonds of
the intermediate zirconacyclopropene complex, Cp2Zr[(MeC6H4)-
CC(C6H4CF3)], toward insertion, but regioselectivity in the
orientation of the second equiv of alkyne as it inserts. For
Heating a solution of the tetraphenyl-substituted zircona-
cyclopentadiene Cp2ZrC4Ph4 at 150 °C in the presence of an
excess (10 equiv) of 1d allowed observation of new zircona-
1
cyclopentadiene products (by H and 19F{1H} NMR spectros-
copy), and PhCtCPh was liberated (Scheme 4). This reaction
was very slow at this temperature, taking over 24 h to go to
completion. After 1 h of heating two new products were
observed and less than 5% of the Cp2ZrC4Ph4 had been
(54) Buchwald, S. L.; Nielsen, R. B. Chem. ReV. 1988, 88, 1047-1058.
(55) Buchwald, S. L.; Nielsen, R. B. J. Am. Chem. Soc. 1989, 111, 2870-
2874.
(56) Takahashi, T.; Kotora, M.; Hara, R.; Xi, Z. Bull. Chem. Soc. Jpn. 1999,
72, 2591-2602.
(57) Nugent, W. A.; Thorn, D. L.; Harlow, R. L. J. Am. Chem. Soc. 1987, 109,
2788-2796.
(59) Quntar, A. A. A.; Srebnik, M. Org. Lett. 2001, 3, 1379-1381.
(60) Hara, R.; Xi, Z.; Kotora, M.; Xi, C.; Takahashi, T. Chem. Lett. 1996, 1003-
1004.
(58) Warner, B. P.; Davis, W. M.; Buchwald, S. L. J. Am. Chem. Soc. 1994,
116, 5471-5472.
9
4202 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
Scheme 4
consumed. These products each exhibit a single resonance in
the 19F{1H} NMR spectrum, one with a chemical shift (δ 0.77)
similar to that of 2d(ââ) and one with a chemical shift (δ 1.22)
that is similar to one of the resonances of 2d(Râ). On the basis
of such comparisons, these two species appear to be the products
of mixed coupling between diphenylethyne and 1d (Scheme 4,
where the â isomer has the p-trifluoromethylphenyl substituent
in the 3-position of the zirconacyclopentadiene and the R-isomer
contains the p-trifluoromethylphenyl substituent in the 2-posi-
tion). The initial ratio of products (70% â isomer) appears to
be kinetically determined, because continued heating eventually
generates a thermodynamic product mixture of these two
monosubstituted isomers, in which the R and the â isomers are
present in approximately a 1:1 ratio. Also observed in the
product mixture are the disubstituted products 2d(ââ) and
2d(Râ). Further heating over a total of 72 h provides a product
mixture containing the isomers 2d(Râ), 2d(ââ), and 2d(RR);
the thermodynamic product distribution of disubstituted products
is 23% RR: 59% Râ: 18% ââ, as measured from both 1H and
19F{1H} NMR spectroscopy. This is close to the 25:50:25
statistical product distribution expected in the absence of any
regioselectivity; however, it is notable that the RR product,
which is kinetically disfavored, is actually thermodynamically
slightly favored over the ââ isomer. The observation of some
selectivity in the initial formation of the monosubstituted R and
â isomers in this reaction indicates that mechanism A is a poor
description of the reaction mechanism, because it predicts that
the second alkyne inserts with no selectivity. In contrast, some
selectivity is observed even at elevated temperatures. This
argument assumes that a dissociative mechanism is operating
(vide infra).
and 2d(Râ) in the presence of an excess of an alkyne, to displace
1 equiv of (p-CF3C6H4)CtC(p-CH3C6H4) and form a new
zirconacyclopentadiene. Mechanism A predicts that the single,
transient zirconacyclopropene generated by heating a mixture
of 2d(Râ) and 2d(ââ) should produce zirconacyclopentadienes
in which the p-CF3C6H4 group is in the â position, whereas
mechanism B predicts that the p-CF3C6H4 group should be
equally distributed between the R and â positions of the product.
Heating a benzene-d6 solution of 2d(ââ) and 2d(Râ) at 140 °C
in the presence of an excess of PhCtC(C6F5) for 2 h produced
a 1:1 ratio of two new monosubstituted products, as shown in
1
Scheme 5, along with unreacted 2d(ââ) and 2d(Râ). The H
and 19F{1H} NMR spectra of the reaction solution indicate the
formation of two products which exhibit chemical shifts for their
p-CF3 groups (δ 0.92, 0.42) that are consistent with those
previously observed for R-C6H4CF3 and â-C6H4CF3 isomers,
respectively. Resonances for the two C6F5 groups are nearly
overlapping, indicating that these substituents occupy the â
positions in both isomers. These results provide further support
that mechanism B is operative in the formation of 2d(ââ) and
2d(Râ), and that the strongest electronic effect is exhibited by
the inserting alkyne. Continued heating of the reaction mixture
for 24 h produces a thermodynamic product mixture of the
monosubstituted isomers, in which one isomer is slightly favored
(58% R-C6H4CF3: 42% â-C6H4CF3), and also results in the
formation of the disubstituted products 2c(ââ) and 2c(Râ), as
expected. Experiments performed with varying amounts of
excess alkyne revealed no significant difference in reaction rates,
which confirms that this reaction occurs by a dissociative
mechanism, rather than by an associative mechanism.
The absence of RR isomers in the coupling of 1a-c can
therefore be rationalized by mechanism B in Scheme 3.
However, in the case of the pentafluorophenyl-substituted
A second way to distinguish between the limiting mechanisms
A and B of Scheme 3 involves heating a mixture of 2d(ââ)
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4203

A R T I C L E S
Johnson et al.
Scheme 5
Scheme 6
alkynes, an additional effect must be operating, because the ââ
coupling product is kinetically favored over the Râ product.
This could be an electronic effect that is not observed with the
p-trifluoromethylphenyl group because it is less electron with-
drawing than the pentafluorophenyl substituent; however, it is
possible that the steric bulk of the pentafluorophenyl group is
also important in the observed regioselectivity.61-64 Further
mechanistic information could be obtained from reversible
coupling experiments with 2c(ââ); however, heating this zir-
conacyclopentadiene at 150 °C provided no evidence that this
coupling was reversible, either via alkyne exchange reactions
or by the formation of 2c(Râ). It was therefore also impossible
to determine the thermodynamic product mixture for the
coupling of 1c, or to determine if the RR isomer is a viable
thermodynamic product despite being kinetically disfavored (as
was observed in the coupling of 1d).
the bis(pentafluorophenyl)-substituted 1,6-heptadiyne (6) was
subjected to coupling conditions with the zirconocene synthon
Cp2Zr(py)(Me3SiCtCSiMe3). This resulted in formation of
zirconacyclopentadiene 7, as shown in Scheme 6. Crystals of 7
suitable for X-ray diffraction were obtained by slow evaporation
of a toluene solution. The solid-state molecular structure (Figure
3) clearly demonstrates that it is possible to couple alkynes with
pentafluorophenyl groups in the two R positions of the zircona-
cyclopentadiene. The structure has noncrystallographic pseudo-
mirror symmetry, such that the mirror plane lies perpendicular
to the zirconacyclopentadiene ring. The pentafluorophenyl rings
form angles of 67.2° and 64.8° to the mean plane of the four
carbon atoms of the zirconacyclopentadiene. The pentafluoro-
phenyl ring ipso-carbons C18 and C24 lie out of the mean plane
of the zirconacyclopentadiene by 0.29 and 0.22 Å, respectively.
This distortion may be due to steric interactions between the
pentafluorophenyl rings and the cyclopentadienyl ligands;
however, other aspects of the structure, such as the Zr1-C17
and Zr1-C11 distances of 2.245(4) and 2.241(4) Å, respectively,
are not significantly different from comparable parameters in
other zirconacyclopentadienes.
To determine if it is possible to overcome the regioselective
effect of the pentafluorophenyl group using a linking group,
(61) Barclay, T. M.; Beer, L.; Cordes, A. W.; Oakley, R. T.; Preuss, K. E.;
Taylor, N. J.; Reed, R. W. Chem. Commun. 1999, 531-532.
(62) Crocker, C.; Goodfellow, R. J.; Gimeno, J.; Uson, R. J. Chem. Soc., Dalton
Trans. 1977, 1448-1452.
DFT Calculations. The coupling of alkynes by zirconium
has been assumed to occur via an intermediate or transition state
in which the two alkynes are activated to an equal extent by
(63) Fenton, D. E.; Massey, A. G.; Jolley, K. W.; Sutcliffe, L. H. Chem.
Commun. 1967, 1097-1098.
(64) AnTorrena, G.; Palacio, F.; Davies, J. E.; Hartley, M.; Rawson, J. M.; Smith,
J. N. B.; Steiner, A. Chem. Commun. 1999, 1393-1394.
9
4204 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
weakly interacting with the zirconium center. The other
optimized structures demonstrate that the subsequent formation
of the Câ-Câ bond occurs via a bis-alkyne complex in which
both alkynes are equally activated by the zirconium center;
however, this bis-alkyne complex is neither an intermediate nor
a transition state on the reaction pathway. The weakly bound
alkyne in the calculated transition state has a CtC bond distance
of 1.217 Å, and is not strongly activated. Conversely, the
strongly bound alkyne has a C-C distance of 1.315 Å, which
is the appropriate distance for a double bond. The near linearity
of the weakly bound alkyne is also evidence of the negligible
back-bonding from the zirconium to this moiety in the calculated
transition state. It is notable that for both alkyne ligands in the
transition state, the Zr-C distance to the carbons on the outside
of the Cp2Zr wedge (CR in the product) are shorter than the
adjacent Zr-Câ distances by ∼0.06 Å.
A similar approach was taken in modeling the coupling of
PhCtC(C6F5) by Cp2Zr. Starting from the optimized ground-
state structure of the zirconacyclopentadiene 2c(ââ), a coordinate
search was performed with the Câ-Câ bond length being
increased in small increments, and single constraint geometry
optimizations were performed at each point. The approximate
transition state structure for alkyne coupling, determined in this
manner, is shown in Figure 5. This transition state is qualitatively
similar to that shown in Figure 4 for the coupling of MeCtCMe,
in that one alkyne is coordinated to the metal and strongly
activated while the second alkyne is in the process of coordinat-
ing to the metal center. The differences between the two Zr-C
bond lengths for each alkyne moiety are even larger in this
transition state than those shown in Figure 4. However, this
may be the result of a somewhat later transition state for Câ-
Câ formation in this case.
Figure 3. ORTEP depiction of the solid-state molecular structure of 7.
Hydrogen atoms and the cocrystallized toluene molecule are omitted for
clarity. Selected bond lengths(Å), bond angles(°) and dihedral angles(°):
Zr1-C11, 2.245(4); Zr1-C17, 2.241(4); C11-C12, 1.341(6); C12-C16,
1.516(5); C16-C17, 1.347(5); C17-C24, 1.484; C11-C18, 1.481(6); Zr1-
C11-C18, 135.3(3); Zr1-C17-C24, 138.4(3); Cp(centroid)-Zr-Cp(cen-
troid), 131.74(2), C12-C16-C17-C24, 170.4(4); C16-C12-C11-C18,
-167.0(4), C16-C17-C24-C25, 120.2(4); C12-C11-C18-C19,
-125.6(4).
The coupling of alkyne 1d by zirconocene discussed in the
previous section demonstrated the importance of an electronic
effect in the regioselectivity of this reaction. Interestingly, this
effect appears to be exerted mainly by the second alkyne, which
is not strongly bound to zirconium in the transition state. A
possible interpretation of this effect is based on a preferred
electrostatic interaction between the zirconium center and the
more electron-rich carbon of the CtC unit (the one bearing
the p-CH3C6H4 group). Alternatively, it may be that increased
molecular orbital overlap between a filled, zirconium-based a1
orbital occurs with the LUMO of the alkyne via direct interaction
with the carbon bound to the p-CH3C6H4 group (where the
LUMO has a stronger contribution).
Figure 4. Calculated transition state for the coupling of MeCtCMe by
Cp2Zr with selected bond lengths and interatomic distances. Cyclopenta-
dienyl hydrogen atoms are omitted for clarity.
the metal center prior to zirconacyclopentadiene formation;56
however, the mechanistic studies utilizing alkyne 1d indicate
that this is not true. To provide further evidence that an
unsymmetrical transition state is relevant in the coupling of
alkynes by zirconocene, DFT calculations were performed on
the Câ-Câ bond cleavages of Cp2ZrC4Me4 and 2c(ââ) using
the B3/LYP level of theory and the LACVP* basis set. A
transition state for Câ-Câ bond cleavage was found by
optimizing the ground state of the zirconacyclopentadiene, and
then performing a coordinate search in which the Câ-Câ bond
was constrained to longer lengths in small increments, with the
structure being optimized with this single constraint at each
point. From these data we determined that the transition state
occurred near a Câ-Câ separation of 3.3 Å. A QST (quadratic
synchronous transit)-guided search then located a transition state
very similar to that found using the coordinate search. This
transition state is shown in Figure 4. The calculated transition
state for the coupling of two alkynes by the zirconocene
fragment contains one alkyne that is strongly activated with
normal Zr-C bond lengths, whereas the second alkyne is only
In the coupling of fluoroaryl-substituted alkynes 1a-1c both
the strongly and weakly bound alkynes in the proposed transition
state appear to affect the regioselectivities, as suggested by the
very high ratio of ââ/Râ isomers (Scheme 1). In these cases,
the insertion into the zirconacyclopropene intermediate prefer-
entially occurs into the Zr-C bond associated with the per-
fluoroaryl group. One explanation for the additional selectivity
observed with pentafluorophenyl substituents is based on the
frontier orbital interactions between the Cp2Zr fragment and the
alkyne moieties. The overlap between the zirconium-based
orbital of a1 symmetry and the π* LUMO of the alkyne moiety
should be greater in the ââ transition state than the Râ transition
state (Figure 5), because the contribution to the π* LUMO is
larger on the side of the alkyne which bears the less electron-
withdrawing substituent. A similar argument could be made for
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4205

A R T I C L E S
Johnson et al.
Figure 5. Calculated transition state of the coupling of PhCtC(C6F5) by Cp2Zr with selected bond lengths and depictions of selected orbital overlaps and
steric interactions in the postulated transition states for the formation of 2c(ââ) and 2c(Râ). All hydrogen atoms are omitted for clarity.
interactions. This is probably due to the geometrical require-
ments of alkyne binding to the transition metal. Although it is
difficult to determine the importance of this π-π interaction,
it provides an alternative explanation for the â-directing effect
of the pentafluorophenyl group. In fact, the small energy
difference between the transition states that lead to the ââ and
Râ coupled products for 1a-c makes it difficult to ascertain
what the most important effect is in the regioselectivity of this
coupling reaction.
Zirconocene Coupling Routes to Fluorinated Polymers.
Preliminary investigations indicate that the regioselective zir-
conocene coupling of pentafluorophenylalkynes may be useful
Figure 6. View of the stacking of the pentafluorophenyl rings in the
calculated transition state for the coupling of PhCtC(C6F5) by Cp2Zr.
Selected distances (Å) under 3.4 Å: F47-C50, 3.267; F46-F52, 3.267;
F55-C41, 3.172; F55-C44, 3.325; F58-C40, 3.369; F49-F58, 3.346;
C27-C28, 3.283; C54-C40, 3.379; C54-C43, 3.393; C42-C28, 3.332.
in the syntheses of pentafluorophenyl-substituted oligomers and
polymers. Diyne 8, synthesized by the method outlined in
Scheme 7, is moderately soluble in dichloromethane and toluene,
and soluble in THF. The zirconocene coupling of 8 in THF
afforded insoluble zirconium-containing polymers. Although this
polymer is depicted in Scheme 7 as having solely ââ-C6F5
substitution of the zirconacyclopentadiene ring, studies with
model alkyne 1c suggest that up to 15% of these linkages are
actually Râ-substituted. However, given the likely conjugation
lengths for these polymers,68 it is expected that the electronic
properties should be dominated by regioselectively coupled
segments. This metal-containing species reacted with HCl and
S2Cl2 to produce butadiene (9) and thiophene (10) polymers,
respectively. Polymers 9 and 10 are slightly to moderately
soluble in organic solvents such as toluene, dichloromethane,
and THF.
the orientation of the second weakly bound alkyne moiety. The
difference in Zr-C bond lengths to the two adjacent carbons
of each alkyne could also play an important role in orbital
overlap; both central Zr-C bond lengths in the transition state
are longer than the adjacent Zr-C bond lengths for both alkynes.
Also, a steric effect could be important here, because the
movement of the bound alkyne across the Cp2Zr wedge will
increase the steric interactions between the cyclopentadienyl
ligands and the ortho fluorine atoms of the fluorinated aryl
groups.61-64 These orbital overlaps and steric interactions for
the transition states which lead to 2c(ââ) and 2c(Râ) are depicted
in Figure 5.
The colors of polymers 9 and 10 are orange and yellow,
respectively, and the UV-vis spectra of these polymers exhibit
λmax values of 382 (9) and 340 (10) nm. The lower λmax value
for polymer 10 is opposite the trend expected for thiophenylenes
vs butadienephenylenes,25 and indicates that the perfluorophenyl
pendant groups disrupt the conjugation of 10 more significantly.
An additional feature of the calculated transition state for
PhCtC(C6F5)-coupling is the near parallel arrangement of the
pentafluorophenyl rings (Figure 5). In this modeled transition
state, there are many short contacts between the carbon atoms
and fluorine atoms of these two aromatic rings. This is illustrated
from a different viewpoint in Figure 6, which also provides
interatomic distances that are less than 3.4 Å. These distances
are comparable to those observed in structures with attractive
π-π interactions;65-67 however, the exact stacking of the rings
is distorted compared to other commonly observed π-π
(65) Adams, N.; Cowley, A. R.; Dubberley, S. R.; Sealey, A. J.; Skinner, M. E.
G.; Mountford, P. Chem. Commun. 2001, 2738-2739.
(66) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525-
5534.
(67) Lorenzo, S.; Lewis, G. R.; Dance, I. New J. Chem. 2000, 24, 295-304.
(68) Mu¨llen, K.; Wegner, G. Electronic Materials: The Oligomer Approach;
Wiley-VCH: New York, 1998.
9
4206 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
Scheme 7
Table 1. Crystallographic Data for the Compounds 2c(ââ), 4c, and 7
4c
2c(ââ)
7
empirical formula
formula weight
crystal color, habit
crystal dimensions
crystal system
a, Å
ZrF10C38H20O2
789.78
orange, plate
0.51 × 0.10 × 0.10 mm
monoclinic
9.515(1)
24.819(3)
16.384(2)
90°
103.597(2)°
90°
ZrF10C28H16(C7H8)0.5
679.71
yellow, bladelike
0.35 × 0.16 × 0.10 mm
monoclinic
7.9714(2)
34.7343(9)
9.7619(1)
90°
C31H16F10S
610.51
yellow, block
0.31 × 0.20 × 0.10 mm
triclinic
12.5044(8)
13.6022(8)
16.7412(10)
79.390(1)°
68.093(1)°
86.523(2)°
2596.6(3)
P1h (#2)
b, Å
c, Å
R, deg
â, deg
98.123(1)°
90°
2675.8(1)
γ, deg
V, Å3
3761.0(7)
P21/c (#14)
Q
space group
Z value
Dcalc
F000
µ(MoKR)
radiation
temperature
scan type
scan rate (s/frame)
2θmax
total no. of reflns
no. of unique reflns
transmission factors
function minimized
no. with I g nσ(I)
no. variables
reflection/parameter ratio
residuals: R; Rw; rall
GOF
P21/c (#14)
4
4
1.395 g/cm3
1576
1.687 g/cm3
1356
1.562 g/cm3
1232
3.72 cm-1
MoKR (l ) 0.71069 Å)
-99 °C
5.02 cm-1
MoKR (l ) 0.71069 Å)
2.18 cm-1
MoKR (l ) 0.71069 Å)
-86.0 °C
ω (0.3° per frame)
10.0
49.4°
ω (0.3° per frame)
10.0
49.4°
ω (0.3° per frame)
10.0
52.3 °
16962
11980
14234
2236 (Rint ) 0.032)
0.81-0.97
ω (|Fo| - |Fc|)2
2414 (n ) 2)
341
3175(Rint ) 0.045)
0.54-0.97
ω (|Fo| - |Fc|)2
3271 (n ) 3)
381
8876 (Rint ) 0.032)
0.82-0.96
ω (|Fo| - |Fc|)2
4321 (n ) 3)
757
7.08
8.59
5.71
0.049; 0.043; 0.111
1.21
0.039;0.050,0.060
1.55
0.034; 0.032; 0.094
1.04
max ∆/σ
0.05
0.4, -0.33
0.00
0.47,-0.57
0.00
0.28, -0.23
residual density, e-/Å3
Table 2. Optical and Molecular Weight Data for Polymers 9 and
10
The molecular weights of these polymers (Mw ) 9260 for
polymer 9; 11840 for polymer 10; by gel permeation chroma-
tography with polystyrene standards) are probably limited by
the low solubility of the zirconocene-based polymer precursor
under the reaction conditions employed. The properties of these
polymers are summarized in Table 2, and the absorption and
emission spectra are shown in Figure 7.
λmax
(nm)
λedge
(nm)
λem
(nm)a
E
g
polymer
M /M
(eV)
Φb
w
n
9
10
9260/4860
11840/5830
382
340
478
438
485
449
2.59
2.76
0.0122
0.0108
a Polymers 9 and 10 were excited at 375 and 340 nm, respectively.
b Quantum yields for the polymers were obtained by measuring the relative
integrated fluorescence intensities compared to a 9,10-diphenylanthracene
standard in dichloromethane.
Conclusions
Despite the synthetic utility of the zirconocene coupling of
alkynes, there are few studies on the influence of electronic
factors on the regioselectivity of this reaction, and likewise little
insight into the reaction mechanism. This investigation into the
coupling of fluoroaryl substituted alkynes has shown that these
electron-withdrawing substituents preferentially adopt the â-
position of the resultant zirconacyclopentadiene. This study also
demonstrates that there is a significant electronic effect in the
regioselectivity of this coupling reaction, and that the transition
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4207

A R T I C L E S
Johnson et al.
excitation source and a Hamamatsu R928P photomultiplier tube
operating at -900 Vdc. Data were collected on deoxygenated solutions
(CH2Cl2 for standards and samples) having an optical density of 0.08-
0.11 (1.0 cm path length) at the excitation wavelength. The excitation
energies for photoluminescence spectra were 10 nm lower than the
absorption maximum, and excitation spectra were acquired at the
emission maximum. No additional corrections were applied. All
quantum yield measurements were carried out at a single excitation
wavelength (350 nm). Quantum yield (Φ) values are reported relative
to a solution of 9,10-diphenylanthracene (Φ ) 0.90). Multiple
measurements on each sample indicated a precision of 10% for the
reported values of Φ.
(p-C6HF4)C≡C(CH2)4CH3 (1a). To a mixture of 1-bromo-2,3,5,6-
tetrafluorobenzene (3.72 g, 16.2 mmol), Pd(PPh3)4 (0.502 g, 0.44 mmol),
CuI (0.408 g, 2.1 mmol) and diisopropylamine (15 mL) in toluene (30
mL) was added 1-heptyne (1.86 g, 19.4 mmol). The reaction mixture
was stirred for 10 h under nitrogen at 95 °C and then the solvent was
removed under reduced pressure. The residue was dissolved in hexane
(200 mL) and purified by column chromatography on basic alumina
Figure 7. UV-vis and photoluminescent (PL) spectra of polymers 9 (UV-
vis: bold solid line; PL: bold dot line) and 10 (UV-vis: thin solid line;
PL thin dot line).
1
using hexane as an eluent to give solid 1a in 77% yield (1.234 g). H
NMR (C6D6, 400 MHz, 293 K): δ 6.24 (m, 1H, C6HF4), 2.14 (t, J )
9.3 Hz, 2H, CH2(CH2)3CH3), 1.37 (m, 2H, CH2CH2(CH2)2CH3), 1.23
(m, 4H, (CH2)2(CH2)2CH3), 0.80 (t, J ) 9.4 Hz, 3H, (CH2)4CH3).
13C{1H} NMR (C6D6, 100.6 MHz): δ 148.5 (m, CC4F4CH), 106.2 (m,
CC4F4CH), 105.3 (t, J ) 25 Hz, CC4F4CH), 104.7 (t, -CtC-), 66.4
(t, -CtC-), 31.5 (CH2(CH2)3CH3), 28.2 (CH2CH2(CH2)2CH3), 22.9
((CH2)2CH2CH2CH3), 19.8 ((CH2)3CH2CH3), 14.1 ((CH2)4CH3). 19F{1H}
NMR (C6D6, 376.5 MHz): δ -90.5 (m), -92.0 (m). EIMS showed
M+ ) 244. HRMS Calcd for C13H12F4, 244.0875; Found, 244.0870.
(p-C6HF4)C≡CCH3 (1b). To a mixture of propyne (1.90 g, 48
mmol) and 1-bromo-2,3,5,6-tetrafluorobenzene (2.512 g, 11 mmol) in
toluene (21 mL) were added Pd(PPh3)4 (0.347 g, 0.30 mmol), CuI (0.288
g, 1.52 mmol) and diisopropylamine (9 mL). The reaction mixture was
stirred for 24 h under an inert atmosphere at 80 °C. The solution was
then filtered and purified by chromatography on basic alumina using
hexane as an eluent. After the solvent was removed, the remaining
solid was sublimed under vacuum to provide 1.24 g of a pale yellow
state in this reaction does not contain two equally activated
alkyne moieties. The nature of the transition state was further
examined by DFT calculations, which confirmed that this
coupling reaction proceeds via a transition state in which one
alkyne is strongly bound to the zirconium center and the other
alkyne is only weakly bound. Further mechanistic insight into
these reactions was obtained from the previously unreported
reversible coupling of some aryl substituted alkynes at 150 °C.
This further understanding of the regioselectivity and mechanism
of this coupling reaction increases the scope of this reaction in
synthesis. For example, the â-directing effect of the fluoroaryl
substituents was utilized for the facile synthesis of new
conjugated polymers bearing electron-withdrawing penta-
fluorophenyl groups. The application of these polymers as
electron-transport materials is currently under study.
1
crystalline solid (60% yield). H NMR (C6D6, 400 MHz): δ 6.15 (m,
1H, C6HF4), 1.51 (s, 3H, CH3). 13C{1H} NMR (C6D6, 100.6 MHz): δ
147.2 (m, CC4F4CH), 106.1 (t, J ) 10, CC4F4CH), 105.4 (t, J ) 20
Hz, CC4F4CH), 100.3 (t, -CtC-), 66.5 (t, -CtC-), 4.0 (CH3).
19F{1H} NMR (C6D6, 376.5 MHz, 293 K): δ -90.7 (m), -91.9 (m).
EIMS showed M+ ) 188, HRMS Calcd for C9H4F4, 188.0249; Found,
188.0248.
Experimental Section
General Procedures. All manipulations were performed under an
inert atmosphere of nitrogen using either standard Schlenk techniques
or a glovebox. Dry, oxygen-free solvents were employed throughout.
All solvents were distilled from sodium/benzophenone ketyl. Benzene-
d6 was purified by vacuum distillation from Na/K alloy and chloroform-
d1 was distilled from CaH2. Mass spectra were recorded on a Micromass
VG ProSpec instrument (ionization energy: 70 eV). IR spectra were
recorded on a Mattson Galaxy Series 3000 spectrometer, and UV-vis
spectra were obtained with a HP 8452A spectrophotometer. NMR
spectra were recorded on Bruker AMX (300 MHz), Varian Bruker
AMX (400 MHz), or Bruker DRX (500 MHz) spectrometers. All
chemical shifts are reported in ppm units. For 19F{1H} NMR spectra,
C6H5CF3 was used as the external reference at 0.00 ppm. The molecular
weights of the polymers were determined by gel permeation chroma-
tography (GPC; Waters R401 Differential Refractometer Detector;
Waters 501 HPLC Pump) with THF as an eluent and polystyrene
standards.
Zirconocene dichloride (Strem or Boulder Scientific) and n-butyl-
lithium solution (Aldrich) were used as received. The compound 1,4-
diethynylbenzene was prepared from 1,4-bis(trimethylsilylethynyl)-
benzene by standard procedures,69 and Cp2Zr(py)(Me3SiCtCSiMe3)
was prepared by a modification of the original synthesis.23 Emission
spectra were collected by using an Instrument SA/Jobin Yvon-Spex
Fluoromax photon-counting fluorimeter equipped with a Xe arc lamp
(C6F5)C≡CPh (1c). To a mixture of phenylacetylene (1.04 g, 10
mmol) and pentafluoro-iodobenzene (3.0 g, 10 mmol) in toluene (30
mL) were added Pd(PPh3)4 (0.346 g, 0.30 mmol), CuI (0.180 g, 0.947
mmol) and diisopropylamine (9 mL). The reaction mixture was stirred
for 24 h under an inert atmosphere at 80 °C. The solution was filtered
and evaporated to dryness. Sublimation at 70 °C under vacuum (1 ×
1
10-2 mmHg) provided 1.92 g of white solid (70% yield). H NMR
(CDCl3, 400 MHz, 293 K): δ 7.62 (d, 2H, ArH), 7.43(m, 3H, ArH).
13C{1H} NMR (CDCl3, 100.6 MHz): δ 147.2 (d, J ) 25 Hz, C5F5C),
141.3 (d, J ) 25 Hz, C5F5C), 137.5 (d, J ) 25 Hz, C5F5C), 131.7 (s,
ArC-H), 129.5 (s, p-ArCH), 128.4 (s, ArC-H), 121.5 (s, ArC), 101.5
(s, CtC), 100.3 (m, C5F5C), 72.9 (s, CtC). 19F{1H} NMR (CDCl3,
376.5 MHz, 293 K): δ -19.8 (m), -28.8 (t, J ) 24 Hz). -45.6 (m).
EIMS showed M+ ) 268. HRMS Calcd for C14H5F5, 268.0311; Found,
268.0309.
(p-CF3C6H4)C≡C(p-MeC6H4) (1d). This compound was prepared
by the same method as 1a using 4-trifluoromethyl-iodobenzene and
tolylacetylene. A white solid was obtained after recrystallization from
1
hexane. Yield: 73%. H NMR (C6D6, 400 MHz): δ 7.40 (d, J ) 8.1
Hz, 2H, ArH), 7.20 (d, J ) 8.1 Hz, 2H, ArH), 7.08 (d, J ) 8.1 Hz,
2H, ArH), 6.77 (d, J ) 8.1 Hz, 2H, ArH), 1.92 (s, 3H, CH3). 13C{1H}
NMR (C6D6, 100.6 MHz): δ 139.58 (s, ArC-H), 132.35 (s, ArC-H),
(69) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis;
2nd ed.; John Wiley and Sons: New York, 1991.
9
4208 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
130.31 (s, ArC), 129.99 (s, ArC), 129.93 (s, ArC), 128.27 (t, J ) 28
Hz, CF3Ar-o-H), 125.86 (q, J ) 28 Hz, CF3Ar-m-H), 123.66 (t, J )
272 Hz, CF3ArC), 120.47 (s, ArC), 93.02 (s, CtC), 88.43 (s, CtC),
21.67 (s, CH3). 19F{1H} NMR (C6D6, 376.5 MHz): δ 0.33. EIMS
showed M+ ) 260. HRMS Calcd for C16H11F3, 260.0813; Found,
260.0808.
consisting of three compounds: Unreacted alkyne (∼20%), the ââ
isomer (∼40%) and the Râ isomer (∼40%). Ratios were obtained by
integration of the peaks in the 19F{1H} NMR spectrum and are in
1
agreement with the integration of the methyl peaks in the H NMR
1
spectrum, which are not completely resolved. H NMR (C6D6, 400
MHz): δ 6.3-7.4 (several phenyl-H signals of the different com-
pounds), 5.98 and 5.93 (Cp ∼1:1), 2.04 (CH3 of ââ and Râ isomer),
1.79 (CH3 Râ isomer). 19F{1H} NMR (CDCl3, 376.5 MHz): δ 0.99
(Râ isomer, 20%), 0.73 (Râ isomer, 20%), 0.58 (ââ isomer, 40%),
0.04 (1d 20%). EIMS showed M+ ) 740. HRMS Calcd for C42H32-
F690Zr, 740.1455; Found, 740.1440. HRMS Calcd for C42H32F691Zr,
741.1465; Found, 741.1466. HRMS Calcd for C42H32F692Zr, 742.1459;
Found, 742.1521.
Cp2ZrC4-3,4-(C6HF4)2-2,5-(C5H11)2 (2a). A 100 mL round-bottom
flask was charged with Cp2ZrCl2 (0.300 g, 1.03 mmol) and THF (40
mL). The solution was cooled to -78 °C and n-BuLi (1.2 mL, 1.92
mmol, 1.6 M in hexane) was added. Then a solution of 1a (0.512 g,
2.09 mmol) dissolved in THF (15 mL) was added and the resulting
solution was stirred until the cooling bath reached room temperature
(over 6 h). The solvent was removed and the residue was extracted
with pentane (3 × 15 mL). Yellow crystals (0.43 g, 64% yield) were
Method (b) Cp2Zr(py)(Me3SiCtCSiMe3) (18.8 mg, 0.038 mmol)
and 1d (20 mg, 0.076 mmol) were dissolved in benzene-d6 (0.6 mL)
in an NMR tube. The solution was left overnight at room temperature.
The product mixture was similar to that observed using method (a),
but also included 5% of a third species, which was assigned as the RR
isomer: 19F{1H} (C6D6, 376.5 MHz) δ 1.07 (Râ isomer 45%), 1.05
1
obtained from this pentane solution at a temperature of -40 °C. H
NMR (C6D6, 400 MHz): δ 6.13 (m, 1H, C6HF4), 6.07 (s, 5H, Cp),
2.04 (t, J ) 7.6 Hz, 2H, -CH2(CH2)3CH3), 1.06 (m, 6H, -CH2(CH2)3-
CH3), 0.82 (t, J ) 7.2 Hz, 3H, -CH2(CH2)3CH3). 13C{1H} NMR (C6D6,
100.6 MHz): δ 199.9 (ZrC2), 144.6 (m, CC4F4CH), 122.6 (C6F4H),
120.5 (ZrCdC), 111.3 (Cp), 103.9 (t, J ) 20 Hz, C5F4CH), 39.9
(-CH2(CH2)3CH3), 32.6 (-CH2CH2(CH2)2CH3), 30.5 (-(CH2)2CH2-
CH2CH3), 22.7 (-(CH2)3CH2CH3), 14.1 (-(CH2)4CH3). 19F{1H} NMR
1
(RR isomer, 5%), 0.73 (Râ isomer 45%), 0.58 (ââ isomer, 50%). H
NMR (C6D6, 400 MHz), select peaks: δ 5.98 (s, 50%, ââ isomer, CpH)
and 5.93 (s, 45%, Râ isomer, CpH 45%), 5.88 (s, 5%, RR isomer,
CpH) 2.04 (overlapping s, CH3 of ââ and Râ isomers), 1.79 (s, CH3
Râ isomer), 1.81 (s, RR isomer, CH3).
(C6D6, 376.5 MHz): δ -90.0 (m), -90.4 (m). EIMS showed M+
)
708.
Cp2ZrC4-3,4-(C6HF4)2-2,5-(CH3)2 (2b). The compound was pre-
pared by the same method used for 2a, employing the alkyne 1b. Yellow
crystals were obtained in 72% yield by cooling a saturated pentane
solution to -40 °C. 1H NMR (C6D6, 400 MHz): δ 6.13 (m, 1H, C6HF4),
5.92 (s, 5H, Cp), 1,42 (s, 3H, CH3). 13C{1H} NMR (C6D6, 100.6
MHz): δ 194.6 (ZrC2), 144.6 (m, CC4F4CH), 122.9 (C6F4H), 112.3
(ZrCdC), 111.5 (Cp), 103.8 (t, J ) 25 Hz, C5F4CH), 22.2 (CH3).
19F{1H} NMR (C6D6, 376.5 MHz): δ -89.6 (m), -90.2 (m).
(C5H11)CHC(C6HF4)C(C6HF4)CH(C5H11) (3a). A 100 mL round-
bottom flask was charged with compound 2a (0.259 g, 0.37 mmol)
and THF (20 mL). Then 6 N HCl (4 mL) was added to the flask and
the mixture was stirred for 2 h. The solvent was removed and the residue
was recrystallized from hexane to afford a colorless product in a yield
1
of 78%. H NMR (C6D6, 400 MHz): δ 6.32 (m, 1H, C6HF4), 5.56 (t,
J ) 7.2 Hz, 1H, CH3(CH2)4 CHdC), 1.73 (q, J ) 7.2 Hz, 2H, CH3-
(CH2)3CH2CHdC), 1.01 (m, 6H, CH3(CH2)3CH2CHdC), 0.73 (t, J )
7.2 Hz, 3H, CH3(CH2)3CH2CHdC). 13C{1H} NMR (C6D6, 100.6
MHz): δ 145.1 (m, CC4F4CH), 135.7 ((C5H11)CHdC), 127.6 ((C5H11)-
CHdC), 118.8 (t, CC4F4CH), 105.8 (t, J ) 22 Hz, CC4F4CH), 31.4
(CH3(CH2)3CH2CHdC), 30.2 (CH3(CH2)2CH2CH2CHdC), 28.6 (CH3-
CH2CH2(CH2)2 CHdC), 22.5 (CH3CH2(CH2)3CHdC), 13.9 (CH3(CH2)4-
CHdC). 19F{1H} NMR (C6D6, 376.5 MHz): δ -90.2 (m), -91.9 (m).
MeCHC(C6HF4)C(C6HF4)CHMe (3b). This compound was pre-
pared by the same method used for 3a, from 2b. The pure colorless
product was obtained by recrystallization from pentane in a yield of
Cp2ZrC4(C6F5)2Ph2 (2c(ââ)). Method (a) The compound was
prepared by the same method used for 2a, except that toluene (30 mL)
was used instead of pentane to extract the product from the crude
reaction mixture. The toluene solution was evaporated to dryness and
the remaining solid was rinsed with 5 mL of pentane and dried. This
yellow solid contains ca. 15% of the Râ isomer as an impurity. Yellow
crystals of the ââ isomer were obtained in 75% yield from slow
1
evaporation of a diethyl ether solution at room temperature. H NMR
(C6D6, 400 MHz): δ 6.9-6.95 (m, 4H, C6H5), 6.65-6.75 (m, 6H,
C6H5), 5.92 (s, 10H, Cp). 13C{1H} NMR (C6D6, 125.77 MHz): δ 202.9
(ZrC2), 147.5 (ArC), 145.1 (m, C6F5), 143.2 (m, C6F5), 141.0 (m, C6F5),
136.3 (m, C6F5), 128.5 (ArC-H), 124.9 (ArC-H), 124.5 (ArC-H),
114.1 (ZrCdC), 113.1 (Cp). 19F{1H} NMR (C6D6, 376.5 MHz): δ -4.2
(m, 2F, p-C-F), -11.9 (t, J ) 18.8 Hz, 4F, o-C-F), -28.4 (m, 4F,
m-C-F). Anal. Calcd for C38H20F10Zr: C, 60.23; H, 2.66. Found: C,
58.86; H, 2.38.
Method b) Cp2Zr(py)(Me3SiCtCSiMe3) (0.210 g, 0.42 mmol) and
1c (0.230 g, 0.86 mmol) were separately dissolved in 10 mL of pentane.
The two solutions were combined at 0 °C and the resulting reaction
solution was stirred at this temperature for 6 h. The solution was then
allowed to warm to room temperature and was stirred for 12 h. The
yellow precipitate was isolated by filtration, washed with pentane, and
dried under high vacuum. Yield: 0.163 g (51%). This product contained
∼20% of the Râ-isomer. 1H NMR of Râ isomer (C6D6, 400 MHz): δ
6.95-7.0 (m, 4H, C6H5), 6.80-6.85 (m, 6H, C6H5), 6.01 (s, 10H, Cp).
19F{1H} NMR of Râ isomer (C6D6, 376.5 MHz): δ -3.1 (m, 1F, p-C-
F), -3.8 (m, 1F, p-C-F), -6.3 (t, J ) 22.5 Hz, 2F, o-C-F), -10.4
(t, J ) 22.5 Hz, 2F, o-C-F), -27.3 (m, 2F, m-C-F), -28.5 (m, 2F,
m-C-F).
1
52%. H NMR (C6D6, 400 MHz): δ 6.32 (m, 1H, C6HF4), 5.42 (q, J
) 6.8 Hz, 1H, CH3CHdC), 1.20 (d, 3H, J ) 6.4 Hz, CH3CHdC).
13C{1H} NMR (C6D6, 100,6 MHz): δ 145.1 (m, CC4F4 CH), 129.7
(CH3CHdC), 128.6 (CH3CHdC), 118.3 (t, CC4F4CH), 105.7 (t, J )
23 Hz, CC4F4CH), 15.0 (CH3CHdC). 19F{1H} NMR (C6D6, 376.5
MHz): δ -89.9 (m), -91.9 (m).
PhCHC(C6F5)C(C6F5)CHPh (3c). A 250 mL Schlenk flask was
charged with Cp2ZrCl2 (0.876 g, 3.0 mmol) and THF (60 mL). The
solution was cooled to -78 °C and n-BuLi (3.75 mL, 6.0 mmol, 1.6
M in hexane) was added. Then a solution of 1c (1.608 g, 6.0 mmol)
dissolved in THF (30 mL) was added and the resulting reaction solution
was stirred until the cooling bath reached room temperature (ca. 6 h).
The reaction mixture was then heated for 1 h at 65 °C. The solution
was then treated with 10 mL of aqueous HCl (6 N) at 0 °C, and then
extracted with hexane (2 × 50 mL). The compound was purified by
flash column chromatography and sublimed at 85 °C under 2-10
1
mmHg. Yield: 1.34 g (84%). H NMR (CDCl3, 400 MHz): δ 7.20
(m, 3H, ArH), 6.92 (m, 2H, ArH), 6.65 (s, 1H, CHdC). 13C{1H} NMR
(CDCl3, 100,6 MHz): δ 144.2 (dm, J ) 25 Hz, CC5F5), 141.4 (dm, J
) 25 Hz, C5F5), 138.0 (dm, J ) 25 Hz, C5F5), 135.3 (s, C5H5C), 135.2
(s, C5H5C), 128.3 (s, C5H5C), 128.2 (s, CdC), 127.9 (s, C5H5C), 126.9
(s, CdC), 112.4 (m, C6F5). 19F{1H}NMR (CDCl3, 376.5 MHz): δ
-20.0 (d, J ) 7.1 Hz), -34.3 (t, J ) 3.5 Hz), -42.1 (m). EIMS showed
M+ ) 538 (100%). HRMS Calcd for C28H12F10, 538.0779; Found,
538.0780.
Cp2ZrC4(C6H4Me)2(C6H4CF3)2 (2d(ââ) and 2d(râ)). This com-
pound was prepared as a mixture of the symmetrical (ââ) and the
unsymmetrical (Râ) isomers. Method (a). The compound was prepared
by the same method used for 2a, from the alkyne 1d. Trituration of
the crude reaction mixture with benzene yielded an orange powder
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4209

A R T I C L E S
Johnson et al.
3d(ââ) and 3d(Râ). These compounds were synthesized by the same
method used for 3a, from 2d(ââ) and 2d(Râ). The product was purified
by flash column chromatography (silica/pentane:diethyl ether ) 200:
1). A first fraction enriched with the Râ isomer was obtained, and a
mixed fraction containing the Râ isomer and the symmetrical isomer
onto a water-cooled coldfinger under full vacuum at a bath temperature
of 110 °C, which yielded 1,7-bis(pentafluorophenyl)-1,6-heptadiyne
(4.85 g, 67%) as a white solid (mp 70-71 °C). 1H NMR (chloroform-
3
d, 400 MHz, 295 K): δ 1.97 (p, JHH ) 7.0 Hz, 2H, CH2CH2CH2),
3
2.69 (t, JHH ) 7.0 Hz, 4H, CtCCH2). 19F{1H} NMR (chloroform-d,
1
in a ratio of 1:2 was eluted second. H NMR of Râ isomer (CDCl3,
376.32 Hz, 295 K): δ -74.5 (m, o-C6F5), -91.5 (m, p-C6F5), -99.7
(m, m-C6F5). 13C{1H} (chloroform-d, 100.56 Hz, 295 K): δ 18.8 (s,
CtCCH2), 26.7 (s, CH2CH2CH2), 65.6 (m, C6F5CtC), 100.3 (m, ipso-
C6F5), 102.2 (m, C6F5CtC), 137.6 (dm, m-C6F5), 141.1 (dm, p-C6F5),
147.4 (ddd, o-C6F5). UV-vis: 240 (43 000), 252 (47 000), 274 (sh,
2600). IR (KBr, cm-1) 2249 m (CtC), 1652 m, 1629 m, 1509 s, 1520
s, 1051 s, 990 s, 864 w, 822 w. Anal. Calcd for C19H6F10: C, 53.79;
H, 1.43. Found: C, 53.92; H, 1.46. EI-MS m/z 424.
400 MHz): δ 7.66 (d, J ) 8.3 Hz, 2H, Hf), 7.43 (d, J ) 8.0 Hz, 2H,
He), 7.26 (d, J ) 8.0 Hz, 2H, Hd), 7.21 (d, J ) 7.8 Hz, 2H, Ha), 7.14
(d, J ) 7.8 Hz, 2H, Hb), 6.86 (d, J ) 8.2 Hz, 2H, Hh), 6.83 (d, J ) 8.2
Hz, 2H, Hc), 6.60 (d, J ) 8.0 Hz, 2H, Hg), 6.40 (s, 1H, Hk), 6.15 (s,
e
1H, Hi), 2.41 (s, 3H, CH3m), 2.21 (s, 3H, CH3 ). 19F{1H} NMR of Râ
isomer (CDCl3, 376.5 MHz): δ 0.43 (s, 3F, CF3), 0.21 (s, 3F, CF3).
1H NMR of symmetric ââ isomer (CDCl3, 400 MHz): δ 7.67 (d, J )
8.0 Hz, 4H, Hd), 7.45 (d, J ) 8.0 Hz, 4H, Hc), 6.88 (d, J ) 8.1 Hz,
4H, Ha), 6.62 (d, J ) 8.1 Hz, 4H, Hb), 6.26 (s, 2H, He), 2.23 (s, 6H,
Hf). 13C{1H} NMR ââ isomer (CDCl3, 125.76 MHz): δ 143.59, 142.78,
137.19 (Cipso-CH3), 133.46, 132.19 (C-He), 130.87 (C-Hc), 129.35
(C-Hb), 128.80 (C-Ha), 125.84 (q, J ) 4 Hz, C-Hd), 123.15 (t, J )
342 Hz, CF3), 21.07 (C-Hf3). 13C{1H} NMR Râ isomer (CDCl3, 125.76
MHz): δ 124.9 (C-Hd), 125.2 (C-Hf), 125.8 and 125.8 (C-Hh), 125.9,
125.9 and 125.9 (C-Hc, C-Hg, C-Hi), 130.0 (C-Ha), 130.1 (C-Hb),
130.2 (C-He), 130.7 (C-Hk). 19F{1H} NMR of ââ isomer (CDCl3,
376.5 MHz): δ 0.43 (s, 6F, CF3). Resonances were assigned using
inverse C-H correlation (1 bond HMQC) and a NOESY experiment.
EIMS of the mixture showed M+ ) 522. HRMS Calcd for C32H24F6,
522.1782; Found, 522.1778.
Cp2ZrC4(2,5-C6F5)(CH2)3 (7). The two solids 1,7-bis(pentafluoro-
phenyl)-1,6-heptadiyne (0.976 g, 2.30 mmol) and Cp2Zr(py)(Me3SiCt
CSiMe3) (1.08 g, 1 equiv) were combined in a 250 mL Schlenk flask
under a nitrogen atmosphere and cooled to -78 °C in a dry ice bath.
Hexanes (40 mL) was added via cannula transfer and the solution was
stirred. The dry ice bath was removed and the solution was allowed to
warm to room temperature, which resulted in the formation of a yellow
precipitate. After 1 h at room temperature, the solution was removed
under vacuum and an additional 20 mL of hexanes was added to the
remaining solid, and then removed by cannula filtration. The yellow
solid was dried under vacuum (1.22 g, 82.4%). Single crystals for X-ray
crystallographic analysis were obtained by slow evaporation of a toluene
1
3
solution. H NMR (benzene-d6, 400 MHz, 295 K): δ 1.20 (p, JHH
)
SC4-2,5-Ph2-3,4-(C6F5)2 (4c). The compound was prepared by a
method analogous to that used for 3c, except that instead of hydrolysis
the reaction mixture was quenched with 1 equiv of S2Cl2 and stirred
for an additional 6 h. Afterward the volatile materials were removed
by vacuum transfer and the residue was extracted into hexane (2 × 50
mL). The pure yellow product was obtained by crystallization from
pentane at -40 °C in 76% yield. 1H NMR (CDCl3, 400 MHz): δ 7.35
(m, 3H, C6H5), 7.28 (m, 2H, C6H5). 13C{1H} NMR (CDCl3, 100,6
MHz): δ 44.5 (dm, J ) 25 Hz, CC5F5), 144.3 (s, SCdC), 141.1 (dm,
J ) 25 Hz, CC5F5), 139.2 (dm, J ) 22 Hz, CC5F5), 132.3 (s, CC5H5),
128.9 (s, CC5H5), 128.8 (s, CC5H5), 128.2 (s, CC5H5), 122.6 (s, SCdC),
109.9 (m, C6F5). 19F{1H} NMR (CDCl3, 376.5 MHz): δ -20.3 (d, J
7.1 Hz, 2H, CH2CH2CH2), 2.00 (t, 3JHH ) 7.0 Hz, 4H, CdCCH2), 5.81
(s, 10H, Cp). 19F{1H} NMR (benzene-d6, 376.32 Hz, 295 K): δ -78.8
(m, o-C6F5), -98.0 (t, p-C6F5), -99.7 (m, m-C6F5). 13C{1H} (benzene-
d6, 100.56 Hz, 295 K): δ 21.6 (s, CH2CH2CH2), 36.3 (s, CdCCH2),
111.7 (s, η5-C5H5), 124.1 (t, JCF ) 41 Hz, ipso-C6F5), 133.9 (s, CdC),
138.1 (m, p-C6F5), 138.1 and 142.7 (m, o-C6F5 and m-C6F5), 166.1 (s,
CdC). Anal. Calcd for C29H16F10Zr: C, 53.95; H, 2.50. Found: C,
54.13; H, 2.48.
Heating of Cp2ZrC4Ph4 with 1d. A 1 mL benzene-d6 solution of
of the bis(cyclopentadienyl)1,2,3,4-tetraphenylzirconacylopentadiene
(29 mg, 0.050 mmol) and alkyne 1d (15 mg, 0.056 mmol) were
combined in an NMR tube equipped with a Teflon valve and the tube
was sealed. The tube was then heated in a 150 °C oil bath. The tube
was removed after 1, 15, and 72 h, and the 1H and 19F{1H} NMR spectra
were obtained. After 1 h a 4% conversion of 1d was observed into 2
new products. 19F{1H} NMR after 1 h (benzene-d6, 376.32 Hz, 295
K): δ 0.03 (1d), 0.76 (70%, â), 1.12 (30%, R). Continued heating
provided a thermodynamic mixture of these two new products as well
as the RR, Râ, and ââ isomers of 2d. 19F{1H} NMR after 72 h (benzene-
d6, 376.32 Hz, 295 K): δ 0.03 (1d), 0.57 (2d-ââ), 0.73 (2d-Râ), 0.76
(â), 1.04 (2d-Râ), 1.07 (2d-RR) 1.12 (R). The experiment was repeated
with 10 equiv of 1d. The initial products and rate of reaction remained
the same; however extended heating over 72 h led to a mixture
containing only the RR, Râ and ââ isomers of 2d. 19F{1H} NMR after
72 h (benzene-d6, 376.32 Hz, 295 K): δ 0.03 (1d), 0.57 (18%, 2d-
ââ), 0.73 (59%, 2d-Râ), 1.04 (59%, 2d-Râ), 1.07 (23%, 2d-RR).
Heating of 2d(râ)/2d(ââ) with with 1c. Solid C6F5CtCPh (8.2
mg, 0.030 mmol) was added to a 1 mL benzene-d6 solution of a 1:1
mixture of 2d(Râ) and 2d(ââ), which was prepared by combining
Cp2Zr(Me3SiCtCSiMe3)(py) (10.1 mg, 0.021 mmol) with 1d (11.2
mg, 0.043 mmol). The solution was transferred to an NMR tube
equipped with a Teflon valve and the tube was sealed. The solution
was then heated in a 150 °C oil bath. The tube was removed after 2 h,
and the 1H and 19F{1H} NMR spectra were obtained. Along with
unreacted starting material and a small amount of 2c, two new products
were observed in a 1:1 ratio: Cp2Zr(C4)-2-C6H4Me-3-C6H4CF3-4-C6F5-
5-Ph (labeled â[CF3]â[C6F5]) and Cp2Zr(C4)-2-C6H4CF3-3-C6H4Me-4-
C6F5-5-Ph (labeled R[CF3]â[C6F5]). 19F{1H} NMR, select peaks, after
2 h (benzene-d6, 376.32 Hz, 295 K): δ 0.95 (s, 50%, R[CF3]â[C6F5]),
) 3.5 Hz), -34.0 (t, J ) 3.5 Hz), -42.4 (m). EIMS showed M+
568 (100%). HRMS Calcd for C28H10F10S, 568.0344; Found, 568.0343.
IR (KBr): 1500s, 990s cm-1
)
.
1,2-(CO2Me)2-4,5-(C6F5)2-3,6-(C6H5)2C6 (5c). This compound was
prepared by a method similar to that used for 3c, except that instead of
hydrolysis the reaction mixture was cooled to 0 °C and one equiv of
MeO2CCtCCO2Me and CuCl were added. The mixture was stirred at
room temperature for 2 h and then heated for 8 h at 65 °C. The reaction
mixture was hydrolyzed by adding aqueous HCl (3N, 10 mL). The
product was extracted into hexane (2 × 50 mL) and purified by silica
gel column chromatography using hexane/ether (50/1) as eluting
solvents. Yield: (55%). 1H NMR (CDCl3, 400 MHz): δ 7.28 (m, 3H,
C6H5), 7.14 (m, 2H, C6H5), 3.52 (s, 3H, OCH3). 13C{1H} NMR (CDCl3,
100,6 MHz): δ 167.1 (s, CO2CH3), 143.5 (dm, J ) 25 Hz, CC5F5),
141.7 (s, C6-C6F5), 141.2 (dm, J ) 24 Hz, CC5F5), 136.8 (dm, J ) 25
Hz, CC5 F5), 136.5 (s, C6-C6H5), 134.7 (s, C6-CO2Me), 130.0 (s,
C6H5), 128.4 (s, CC5H5), 128.1 (s, CC5H5), 128.0 (s, C5H5), 112.3 (m,
C6F5), 52.6 (s, CO2CH3). 19F{1H} NMR (CDCl3, 376.5 MHz): δ -20.7
(m), -30.2 (t, 24 Hz), -44.1 (m). EIMS showed M+ ) 678 (100%).
HRMS Calcd for C34H16F10O4, 678.0889; Found, 678.0898.
[(C6F5)CtCCH2]2CH2 (6). A solution of 1,6-heptadiyne (1.567 g,
0.017 mol) and C6F5I (10.0 g, 0.034 mol, 2 equiv) in 200 mL of Et3N
was added to a mixture of solid Pd(PPh3)4 (1.00 g, 0.025 equiv) and
CuI (0.32 g, 0.05 equiv) at room temperature under a nitrogen
atmosphere. The solution was stirred and refluxed for 24 h. The solvent
was then removed under vacuum and the remaining solid was extracted
into 300 mL of pentane and filtered in air. The pentane solution was
then evaporated to dryness and the solid that remained was sublimed
9
4210 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Zirconocene Coupling of Alkynes
A R T I C L E S
0.47 (s, 50%, â[CF3]â[C6F5]), -77.05 (overlapping m, â[CF3]â[C6F5]
and R[CF3]â[C6F5] o-C6F5), -94.65 and -94.95 (m, â[CF3]â[C6F5]
and R[CF3]â[C6F5] p-C6F5), -101.65 and -101.95 (m, â[CF3]â[C6F5]
and R[CF3]â[C6F5] m-C6F5). 1H NMR, select peaks, (benzene-d6,
400 MHz, 295 K): δ 1.76 and 1.99 (s, 1:1, â[CF3]â[C6F5] and
R[CF3]â[C6F5] p-C6H4CH3).
Calculations
Ab initio DFT calculations were performed using the hybrid
functional B3LYP70 method with the Jaguar71 package. The
basis functions and effective core potentials used were the
LACVP* set,72 provided in the Jaguar program. The ground-
state structures were optimized in C2, Cs, and C1 symmetry.
The Câ-Câ bond length was then incrementally increased from
the ground-state structure, and the geometry along each point
in the coordinate search was optimized in C1 symmetry. In the
case of the MeCtCMe coupling, this provided an initial point
to perform a QST guided transition state search; the resulting
structure was not qualitatively significantly different from the
initial search structure. In the case of the PhCtC(C6F5)
coupling, the QST guided search failed to find a transition state,
so the approximate transition state used is that found using the
coordinate search.
1,4-[(C6F5)C≡C]2C6H4 (8). To a mixture of pentafluoroiodobenzene
(11.76 g, 40.0 mmol), Pd(PPh3)4 (0.98 g, 0.84 mmol), CuI (0.20 g, 1.0
mmol) and diisopropylamine (15 mL) in toluene (120 mL) was added
1,4-bisethynylbenzene (2.52 g, 20.0 mmol). The reaction mixture was
stirred for 48 h under nitrogen at 95 °C and then the solvent was
removed under reduced pressure. To this residue was added water (100
mL), and the resulting mixture was extracted with a mixture of hexane/
dichloromethane (1:1, 2 × 100 mL). The solvents were removed and
the residue was purified by silica gel column chromatography using
hexane/ether/methylene chloride (2/1/1) as an eluent. The product was
isolated as a pale yellow solid in 65% yield (6.6 g). 1H NMR (CDCl3,
400 MHz): δ 7.60 (s). 13C{1H} NMR (CDCl3, 124 Hz): δ 147.0 (d,
J ) 25 Hz, C5F5C), 141.5 (d, J ) 25 Hz, C5F5C), 137.8 (d, J ) 25 Hz,
C5F5C), 131.9 (s, ArC-H), 122.7 (s, p-ArC), 100.6 (s, CtC), 100.0
(m, C5F5C), 75.4 (s, CtC). 19F{1H} NMR (chloroform-d1, 376.5
MHz): δ -20.2 (m), -29.8 (t, J ) 24 Hz), -46.0 (m). IR (KBr) 2223
w CtC), 1524 s, 1498 s (aromatic CC), 1116 m (CF), 999 s, 996 m,
836 m, 685 m cm-1. EIMS showed M+ ) 458. HRMS Calcd for
C22H4F10, 458.0153; Found, 458.0149.
Acknowledgment is made to the National Science Founda-
tion for their support of this work. S.Z. is grateful to the Swiss
National Science Foundation and the Novartis Stiftung, vormals
Ciba-Geigy-Jubila¨ums-Stiftung for financial support. Funding
for S.A.J. was provided by the NSERC through a postdoctoral
fellowship. We thank Dr. Fred Hollander and Dr. Allen Oliver
of the UC Berkeley CHEXRAY facility for the X-ray structure
determinations. We also thank Dr. Hyuk Lee, Ben Mork, and
Dr. Patrick Egli for experimental assistance.
[C6H4CHdC(C6F5)C(C6F5)dCH]n (9). The polymer was prepared
by a method similar to that used for compound 3a. After hydrolysis,
an orange solid was precipitated by the addition of methanol to the
solution. The polymer was further purified by precipitation from THF
by the addition of methanol and dried under vacuum. Yield: 80%.
GPC: Mn/Mw ) 4860/9280. 1H NMR (CDCl3, 400 MHz): δ 6.6-7.0
(br, 4 H, C6H4), 6.4-6.6 (br, 2H, CHdC-C6F5). IR (KBr) 1651 m
(CdC conjugated with aryl), 1603 (CdC conjugated with CdC and
aryl), 1500 s, 1498 s (aromatic CdC), 1120 s (CF), 977 s, 940 m, 885
Supporting Information Available: 1H NOESY and C-H
correlation spectra for the two isomers of 3d(ââ) and 3d(Râ).
Tables of crystal data collection and refinement parameters,
coordinates, bond distances, angles, and anisotropic displace-
ment parameters for 2c(ââ), 4c and 7 . Coordinates for DFT
geometry optimizations (66 pages, print/PDF). This material is
contained in many libraries on microfiche which immediately
follows this article in the microfilm version of the journal and
can be ordered from ACS or downloaded from the Internet;
see any current masthead page for ordering information and
Internet access instructions. This material is available free of
charge via the Internet at http://pubs.acs.org.
m, 837 m, 812 m, 551 m cm-1
.
[-C6H4CdC(C6F5)C(C6F5)dCS]n (10). The polymer was pre-
pared by a method analogous to that used for compound 4c. After
work up, a yellow-green solid was isolated from THF/methanol in a
1
yield of 74%. The H NMR spectrum in CDCl3 contained no vinylic
JA0209161
resonances. GPC: Mn/Mw ) 5830/11,840. 1H NMR (CDCl3, 400
MHz): δ 7.16-7.3 (br, C6H4). IR (KBr) 1652 w (CdC conjugated
with aryl), 1521 s, 1496 s (aromatic CdC), 1097 m (CF), 991 s, 918
(70) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(71) Schro¨dinger, Inc.: Portland, OR, 1998.
(72) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
m, 732 m cm-1
.
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4211